SciELO - Scientific Electronic Library Online

Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados


Journal of the Chilean Chemical Society

versión On-line ISSN 0717-9707

J. Chil. Chem. Soc. v.49 n.3 Concepción sep. 2004 


J. Chil. Chem. Soc., 49, N 3 (2004): 251-256



Alexánder Berrocal1,2, Jaime Baeza1, Jaime Rodríguez1,2, Miguel Espinosa2, Juanita Freer1*

1Laboratorio de Recursos Renovables, Facultad de Ciencias Químicas, Universidad de Concepción, Casilla 160-C, Concepción Chile. *E-mail
2 Facultad de Ciencias Forestales, Universidad de Concepción, Casilla 160-C, Concepción Chile

(Received: November 12, 2003 - Accepted: July 9, 2004)


In this current research, the chemical composition of Pinus radiata D. Don trees from different age classes was determined. To achieve this, a hydrolysis with trifluoroacetic acid method was used. Significant differences on chemical composition associated to age were found. Values obtained are necessary to establish feasibility of using this biomass resource at ages in which, sugar contents result both quality and quantity favorable to produce biofuels (bioethanol process) or chemical compounds (biorefinery process).

Keywords: acid hydrolysis, TFA, woods chemical composition, wood sugar, Pinus radiata, biofuels, bioethanol.



OPEC embargo in the seventies, promoted an increase in oil prices and induced countries to develop a self-efficiency plan about fuels. This fact allowed that, in the last 25 years, a great interest on producing ethanol from biomass to become fuel, has been showed.

Global biomass production has been estimated in more than 170 billion tons a year. A 70 % of this comes from forests, making lignocellulosic material, the most important component of this resource. About 50 % of wood is made of cellulose, which makes it the main source of raw material for ethanol production. There are other polysaccharides in wood, but present in lesser proportions (20-25%) and they can also be used for bioethanol production3. Economic analyses have showed that the success in conversion processes from lignocellulosic materials to biofuels is determined by the effective using of all their components: cellulose, hemicellulose and lignin. This, has been defined as biorefinery4-5.

Chile is a forest country, with a surface of 2 037 403 hectares planted6. The main alternative for ethanol production comes from P. radiata wood, by using debris from mill industry, cellulose industry (fine debris) and from forest plantations7.

Prior to the implementation of a great scale bioethanol production project, it is necessary to know the chemical composition of raw material8 and, if a renewable and chemically heterogeneous, as wood, is about to be used, several factors which determine chemical component abundance both in quality and in quantity, must be considered9.

In the specific case of using lignocellulosic material for bioethanol production, is also important to consider sugars composition, because only hexoses are easily fermentable, while pentoses require genetically modified microorganisms to be fermented10.

Among the factors to be considered, forest species, age of tree, genetic selection, geographic region where the tree grows and silvicultural practices for the growing stand, are the most important11. The fact that the macromolecular components are not equally distributed in the cell wall and that their concentration changes from one morphologic region to another is also important9. From all these factors, the first two are the most important and plenty of information about average chemical composition for several species, is available11. However, related to age, this information is scarce and not systematic.

Within a species, there are significant variations on contents of cellulose, hemicellulose and wood lignin11. For Douglas fir, within the same growing ring, cellulose content varies from 40% in the earlywood, until 50% in the latewood, with a lignin decreasing from 28 to 24% under the same conditions12. In Pinus resinosa latewood, both lignin and xylanes content is lower, while the mannanes content is higher, when compared with earlywood5.

Much of this variation is related with specific weight13. In Pinus contorta, a positive correlation between cellulose percentage and wood specific weight was found and, because it there exists an equal relation between age and specific weight, an increase of cellulose percentage along the age, is also found14.

A study performed on Pinus taeda trees, established that the age of the tree is directly related to the content of alpha-cellulose and holocellulose, while the hemicellulose and chemicals content extractable in ethanol-benzene, are inversely proportional to age, but no relation between lignin proportion and tree age was found15. In this case, basic density of wood showed a directly proportional relation to the alpha-cellulose, holocellulose and extractable chemicals, while in the case of hemicellulose, the relation was inversely proportional and when lignin did not show any relation.

In full-grown P. radiata, in Chile, cellulose percentage is lower 45% and that one of extractable chemicals does not reach 2%16, but also in this species, but growing in Australia, it has been informed that the cellulose proportion, increases along age17. In Cupressus lusitanica plantations established in Chile, extractable chemicals content decreases from pith to bark. On the contrary, holocellulose percentage increases in the same direction and the lignin proportion decreases from pith to bark18.

For the specific case of sugar proportion in wood, it has been found that, in P. radiata y P. resinosa, there is a large decrease on galactose, xylose and arabinose content, as also as an increase of mannose from pith to outside in 20 years old trees19.

The aim of this study is to quantify the chemical composition of different age's P. radiata wood. To achieve this, a hydrolysis with trifluoroacetic acid (TFA) modified method, will be used. Because of the interest in using this raw material for obtaining ethanol, the study referring to sugar percentages it is emphasized, standing out those ones easily fermentable (hexoses)


2.1. Sampling.

Wood samples were collected in P. radiata plantations at a location called San Pedro de la Paz, in Concepción, Chile. A selections of stands by means of age classes (1-3; 4-6; 7-10; 11-15; 16-20; 21-25; 26-30 and more than 30 years), was made. Three trees from every one of the stands were randomly selected. From every tree, a round cross section was cut at an average height of 1.3 m, except for the first 3 trees, because they were individuals of less than a meter height. In this case, the wood round cross section was obtained from a portion of the stem located 10-cm above the root.

2.2 Sample processing

Wood round cross sections were air dried during one and half month. Once wood reached humidity content of about 15%, it was chopped. Then, chops were grinded in a Retsch SM1 mill. Samples were sifted to a particle size lower than 0.35 mm (45 mesh) and greater than 0.25 mm (60 mesh), according to norm Tappi 204 om-88. Material was stored in vacuum-sealed plastic bags, to avoid both humidity and fungus. Finally, samples were stores at environment temperature for their later chemical analysis.

2.3 Percentage of wood extractable chemicals determination

Humidity content of every one of the samples was obtained by means of a thermoscale, according to norm Tappi 264 cm-97. Wood was extracted in a soxhlet system for 5 hours, with a solvent recycling of 6 times per hour, using 150 mL of a 1:2 (v/v) ethanol-toluene mixture proportion. After the extraction in ethanol-toluene, samples were filtered Büchner vacuum funnel in order to remove the excess of solvent and then, washed with ethanol to remove toluene remaining. Samples were placed again in the soxhlet device and a new extraction was performed, with 150 mL of ethanol, following the same procedure already described. Once again, samples were transferred to Büchner funnel to remove the solvent excess with vacuum. Both funnel and sample were washed in distilled water in order to remove ethanol. Then, samples were place in a 1000 mL Erlenmeyer flask and 500 of distilled water was added and heated in a hot water bath for 1 hour. Later, samples were filtered in a Büchner funnel and washed with plenty of distilled water.

Extracted wood sample was air dried for several days, being moved with a stick for a homogenous dry and then, stored in a plastic bag and vacuum-sealed for its later analysis. Both organic extracts were mixed in a 250 mL flask and the solvent was partially removed in a rotavapor. Then, it was transferred into a previously tared 50-mL flask. Solvent was eliminated up to complete dryness. The residue was placed in a stove at 105 ± 3 °C for 1 hour and then, was finally was let to cool in a dessicator and weighted with a precision of 0.0001 g. A blank was performed by evaporating 150 mL of solvent in a tared flask. Then, the residue was weighted with a precision of 0.0001 g. the weight of the extract was corrected with the blank, by weight difference.

2.4 Hydrolysis method

A hydrolysis method with trifluoroacetic acid (TFA) was carried out. TFA is a good cellulose solvent. However, this kind of hydrolysis must be optimized depending on the material, which is working with20. Variables modified to optimize hydrolysis were reflux time and temperature of primary hydrolysis, amount of acid added and implementation of an autoclave step after the secondary hydrolysis.

A sample of 100 mg of wood free of extractable chemicals was placed in contact to 8.1 mL of concentrated TFA during 12 hours at environmental temperature. In the first step of hydrolysis, solution in reflux was placed at 97 °C for one hour. In the second step of hydrolysis, TFA was diluted at 80% and it was refluxed at 97 °C for 15 min. After that, TFA was diluted to 30% and the solution was left in reflux at 97 °C for 2 hours. TFA was diluted at 6% and it was autoclaved at 100 °C for 2 hours. Once hydrolysis stage was completed, hydrolyzed elements were separated from residue (lignin) through a crucible number 4, previously tared. Hydrolyzed material was placed in a flask and TFA was evaporated in a rotavapor. Water was added to the flask with the hydrolyzed material and rotavapored several times until pH 4 was reached. Residue of sugars was dissolved in 4mL of HPLC quality water, and analyzed by HPLC using Merck Hitachi equipment with a Differential Refractomer RI-71 detector. A Bio-Rad Aminex HPX-87P column at 85 °C and HPLC degree water as mobile phase was used.

2.5 Statistical analyses

In order to perform chemical analyses of the samples from different age categories, 8 age classes and 3 trees per class were considered. Analyses were performed in triplicate. For every one of the wood chemical components, An ANOVA by means of a quadratic linear regression with 95% of confidence level was carried out in order to identify if whether or not significant differences exist. All 24 samples were considered as the population for the analyses (3 trees per every one of the 8 age classes).


3.1. Wood chemical composition

According to results, it can be observed that the cellulose proportion increased from age class 1-3 years (31.05%) to the age class 21-25 years (42.51%), decreasing in later categories of age (40.51% and 36.36% respectively). On the other hand, hemicellulose percentage decreased from age class 1-3 years to 16-20 years, increasing again at age classes 21-25 and 26-30 years. Percentage of hemicellulose obtained, does not include acetyl groups and glucoronic acids (Table 1).

                * Without including acetyl groups and glucoronic acids
                ** TFA lignin
                *** Soluble extractables in ethanol-toluene

Lignin proportion decreased from age class 1-3 years (38.96%) to the 21-25 years (30.36%), to be increased again in later age classes. An inverse relation between percentage of cellulose and lignin in the different age classes was observed.

Percentage of extractable chemicals obtained, decreased from age class 1-3 years (4.65%) to the 16-20 years (1.74%), increasing in subsequent age classes. Their behavior was similar to that one showed by the proportion of hemicellulose at different age classes.

3.2 Total carbohydrates

Proportion of total carbohydrates tends to increase along age, from age class 1-3 years to the 26-30 years, going from 57% to 65%. At the age class 30 or more, percentage of total carbohydrates decreased to 60%, as a result of the decreasing in cellulose proportion in this age class (Figure 1).

Fig. 1. Percentage of total carbohydrates obtained from P. radiata trees of different age classes, by means of the optimized TFA hydrolysis method.

3.3 Sugar proportion


Glucose is a hexose; results show that its proportion in P. radiata wood, tends to increase with age. In the first years, percentage found is about 40%. Then, the value increased to 50% (11-15 years). Such percentage remains constant up to 30 years. Then, a slight decrease to 45% in the age class of 30 years or more, is observed (Figure 2).

Fig. 2. Glucose percentage found in P. radiata trees of different age classes, by means of the optimized TFA hydrolysis method (•: Average, _ : Tendency).

These values are comparable with those reported by the literature. Juvenile wood contains less glucose than mature wood21. A greater amount of glucose in P. radiata trees from 21 to 25 years when compared with trees from 1 to 5 years, has been also found17.


This sugar is a pentose; its proportion in softwoods is lower than 10%. Percentage of xylose tends to be greater in the first age classes (Figure 3). This is because in juvenile wood, a greater amount of this sugar is found when compared to mature wood21. At the age class 1-3 years, a xylose proportion close to 8% was found. The same one, decreased to 4.5% at the age of 15 years and finally, become to 4% in mature ages (30 years or more).

Fig. 3. Percentage of xylose found in P. radiata trees of different age, by means of the optimized TFA hydrolysis method (•: Average, _ : Tendency).


This sugar is a hexose and it is found in a very low proportion in P. radiata wood. It can be observed that the amount tends to decrease from year 1 to 15 (6% to 2.6% respectively). Then it increases again, reaching a value of 4.5% at 30 years of age (Figure 4).

Fig. 4. Percentage of galactose found in P. radiata trees by means of the optimized TFA hydrolysis method (•: Average, _ : Tendency).

At ages of 1 and 5 years, galactose values found were 10% and 8% respectively. These percentages are associated to the high concentration of juvenile wood present in trees belonging to these age classes21.


Arabinose is a pentose present in low amounts in P. radiata wood (less than 3%). Distribution of arabinose percentage in different age classes, is similar to the one showed by galactose. A decreasing in its proportion from 1-3 years, to 11-15 years (2% and 1% respectively) to subsequently be increased at 30 years of age (1.3%) (Figure 5).

Fig. 5. Percentage of arabinose found in P. radiata trees by means of the optimized TFA hydrolysis method (•: Average, _ : Tendency).


Mannose is a hexose and, in importance, it is the second kind of sugar, because of its proportion in P. radiata wood. According to results, it can be observed that the proportion of mannose in P. radiata wood, tends to increase as the tree goes from 1-3 years of age to 30 years of age (Figure 6). This is due to two main reasons: Juvenile wood presents a lower percentage of mannose when compared with mature wood. On the other hand, trees belonging to lower age classes have a greater percentage of compression wood related to the proportion of normal wood and, their compression wood is characterized by having lower amount of mannose than normal wood21.

Fig. 6. Percentage of mannose found in P. radiata trees by means of the optimized TFA hydrolysis method (•: Average, _ : Tendency).

From analyzed trees it was found that the proportion of mannose increases from 9.5% in the age class 1-3 years, until 14.5% in age class 26-30 years. Later, it tends to decrease in trees older than 30 years. Table 2 indicates average values of glucose, xylose, galactose, arabinose and mannose proportion, found in wood samples of P. radiata of different age classes.

3.4 Proportion of hexoses and pentoses

Percentage of hexoses increased in age class 1-3 years to the age class 21-25, going from 53% to 65%. This percentage remained similar in the next age class and later, decreased to 61.4% at the age class of more than 30 years (Figure 7). On the other side, pentoses exhibited a completely opposite behavior. Their proportion decreased from the age class 1-3 years to the age class 21-25 (9.9% and 4.5% respectively). Percentage of pentoses slightly increased (4.9% y 5.2%) at more mature ages (26-30 and 30 or more years).

Fig. 7. Proportion of hexoses and pentoses found in different age classes P. radiata trees, by means of the optimized TFA hydrolysis method.

The relation in the proportion of hexoses and pentoses is very important if ethanol wants to be obtained from wood, because hexoses are easily fermentable by microorganisms like Saccharomyces cereviseae, while pentoses require genetically modified microorganisms, with all subsequent associated complications.

Table 3 summarizes ANOVA performed for different chemical components of wood. These components will allow establishing if, for every one of them, a statistical difference associated to age exists.

Analysis of variance performed, indicates a significant difference associated to age of all wood chemical components.

Cellulose (%) = -1.158 + 0.943*Glucose (%)1 R2 = 0.993
1 Proportion of glucose present in cellulose

Hemicellulose (%) = -0.028 + 0.521*Glucose (%)2 + 1.002*Xylose (%) + 1.000*Galactose (%) + 0.998*Arabinose (%) + 1.140*Mannose (%) R2 = 1.000
2 Proportion of glucose present in hemicellulose


Chemical composition of P. radiata wood presents variations associated to tree age. These variations are statistically significant, and along the differences found among trees of the same age, show the existent variability found in the wood chemical component proportion. These variations must be taken into account in order to select an optimum material for ethanol production. Low lignin proportion, high concentration of sugars, mainly hexoses because of the easy fermentation, are the main features to be considered. In addition, such material must come from relatively younger age classes, because the success in producing ethanol from lignocellulosic material, depends on the feasibility of establishing low rotation harvest times


Authors wishes to thank FONDECYT (Grant N° 1020161), University of Concepción Research (Grants DIUC 201.023.027-1.0) and to INNOVA BIO BIO - ABENGOA CHILE S.A. (Grant 03-A1-153).


1. Saddler, J.N. and Gregg, D.J. 1998. Ethanol production from for est wastes. In: Bruce, A and Palfreymman, J. W. (Eds.). Forest Product Biotechnology. Taylor and Francis. pp 183-195.         [ Links ]

2. Skog, K and Rosen, H. 1997. United States wood biomass for energy and chemicals: Possible changes in supply, end uses and environmental impacts. Forest Products Journal. 47 (2): 63-69.         [ Links ]

3. Goldstein, I. S. 1992. Recent developments in cellulose hydroly sis by concentrated acids. In 2nd National Symposium on Produc tion of New Proteins and Utilization of Unexplored Resources. October 1985 Lisboa, Portugal. 10 p.         [ Links ]

4. Lee, Y and Mc Caskey, A. 1983. Hemicellulose hydrolysis and fermentation of resultanting pentoses to ethanol. Tappi Journal. 66 (5): 102-107.         [ Links ]

5. Galbe, M. and Zacchi, G. 2002. A review of the production of ethanol from softwood. Applied Microbiological Biotechnology. 59: 618-628.         [ Links ]

6. INFOR. 2002. Estadísticas Forestales 2001. Boletín Estadístico N 79. Instituto Forestal. Santiago, Chile. 145 p.         [ Links ]

7. Schaffeld, G. 1994. El Material Lignocelulósico. En: Cunningham, R y López, G. (Eds): Etanol de Lignocelulósicos. Tecnología y Perspectivas. Programa Iberoamericano de Ciencia y Tecnología para el Desarrollo (CYTED). Universidad de Santiago de Compostela, España. pp 15-34.         [ Links ]

8. Johnson, D. K.; Adam, P.; Ashley, P.; Chum, H.; Deutch, S.; Fennell, J. and Wiselogel, A. 2000. Study of compositional changes in biomass feedstocks upon storage. Changes in Biomass Feed stocks. National Renewable Energy Laboratory. Golden, CO 80401. 20 p         [ Links ]

9. Baeza, J. and Freer, J. 2001. Chemical characterization of wood and its components. In Hon, D.N.S. and Shiraishi, N. (Ed.): Wood and Cellulosic Chemistry. Marcel Dekker. New York, USA. pp 275-384.         [ Links ]

10. Wyman, C. 2001. Twenty years of trials, tribulations and research progress in bioethanol technology. In: Davidson, B.; McMillan, J. and Finkelstein, M. (Eds): Applied Biochemistry and Biotech nology. Proceedings of The Twenty-Second Symposium on Bio technology for Fuels and Chemicals; Held May 7-11 Gatlinburg, Tennessee. Totoa, NJ: Humana Press. pp 5-21.         [ Links ]

11. Harris, J. M. 1981. Wood quality of radiata pine. Appita 35: 211- 215.         [ Links ]

12. Kennedy, L. R. and Warren, W. G. 1969. Within-tree variation in physical and chemical properties of Douglas-fir. Second World Consultation on Tree Breeding FAO/IUFRO Vol 1- pp 393-417.         [ Links ]

13. Zobel, B. 1971. Genetic manipulation of wood of the southern pines including chemical characteristics. Wood Science and Tech nology. 5: 255-271.         [ Links ]

14. Kim, W. J.; Campbell, A. G. and Koch, P. 1989. Chemical varia tion in lodgepole pine with latitude, elevation and diameter class. Forest Products Journal. 39 (3): 7-12.         [ Links ]

15. Mc Millin, C. W. 1968. Chemical composition of loblolly pine wood as related to specific gravity, growth rate and distance from pith. Wood Science and Technology. 2:233-240.         [ Links ]

16. Mansilla, H.; García, R.; Tapia, J.; Durán, H. and Urzúa, S. 1991. Chemical characterization of chilean hardwoods. Wood Science and Technology. 25:145-149.         [ Links ]

17. Uprichard, J. M. and Lloyd, J. A. 1980. Influence of tree age on the chemical composition of radiata pine. New Zealand Journal of Forestry Science. 10 (3): 551-557.         [ Links ]

18. Rodríguez, S y Torres, M. 1995. Estudio de la composición química de la madera de ciprés (Cupressus lusitanica) en relación al fuste. Ciencia e Investigación Forestal. 9(1): 47-56.         [ Links ]

19. Hillis, W. E. 1984. High temperature and chemical effects on wood stability. Wood Science and Technology. 18: 281-293.         [ Links ]

20. Fengel, D. and Wegener, G. 1979. Hydrolysis of polysaccharides with trifluoroacetic acid (TFA) and its application to rapid wood and pulp analysis. In: Brown, R.D. Jr. and Jurasek, L. (Eds.): Hy drolysis of Cellulose: Mechanism of Enzymatic and Acid Cataly sis. Advances in Chemistry Series. 181:145-158.         [ Links ]

21. Saka, S. 2001. Chemical composition and distribution. In: Hon, D.N.S. and Shiraishi, N. (Eds): Wood and Cellulosic Chemistry. Marcel Dekker New York, USA. pp 51-82.         [ Links ]

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons